In the field of optoelectronics,certain types of data may be difficult to accurately annotate,such as high-resolution optoelectronic imaging or imaging in certain special spectral ranges.Weakly supervised learning can...In the field of optoelectronics,certain types of data may be difficult to accurately annotate,such as high-resolution optoelectronic imaging or imaging in certain special spectral ranges.Weakly supervised learning can provide a more reliable approach in these situations.Current popular approaches mainly adopt the classification-based class activation maps(CAM)as initial pseudo labels to solve the task.展开更多
In this article,reference 43 was incorrectly given as‘Erhard,L.C.,Rohrer,V.,Albe,K.,and Deringer,V.L.,Modelling atomic and nanoscale structure in the silicon-oxygen system through active machine learning,https://arxi...In this article,reference 43 was incorrectly given as‘Erhard,L.C.,Rohrer,V.,Albe,K.,and Deringer,V.L.,Modelling atomic and nanoscale structure in the silicon-oxygen system through active machine learning,https://arxiv.org/abs/2309.03587(2023)’and should have read‘Erhard,L.C.,Rohrer,J.,Albe,K.et al.Modelling atomic and nanoscale structure in the silicon–oxygen system through active machine learning.Nat Commun 15,1927(2024).https://doi.org/10.1038/s41467-024-45840-9’.展开更多
Understanding and accurately predicting hydrogen diffusion in materials is challenging due to the complex interactions between hydrogen defects and the crystal lattice.These interactions span large length and time sca...Understanding and accurately predicting hydrogen diffusion in materials is challenging due to the complex interactions between hydrogen defects and the crystal lattice.These interactions span large length and time scales,making them difficult to address with standard ab-initio techniques.This work addresses this challenge by employing accelerated machine learning(ML)molecular dynamics simulations through active learning.We conduct a comparative study of different ML-based interatomic potential schemes,including VASP,MACE,and CHGNet,utilizing various training strategies such as on-the-fly learning,pre-trained universal models,and fine-tuning.By considering different temperatures and concentration regimes,we obtain hydrogen diffusion coefficients and activation energy values which align remarkably well with experimental results,underlining the efficacy and accuracy of ML-assisted methodologies in the context of diffusive dynamics.Particularly,our procedure significantly reduces the computational effort associated with traditional transition state calculations or ad-hoc designed interatomic potentials.The results highlight the limitations of pre-trained universal solutions for defective materials and how they can be improved by fine-tuning.Specifically,fine-tuning the models on a database produced during on-the-fly training of VASP ML force-field allows the retrieval of DFT-level accuracy at a fraction of the computational cost.展开更多
With the initiation of the National Virtual Simulation Experimental Teaching Project in 2018,educational institutions in China have recognized the significance of virtual simulation technology in reforming traditional...With the initiation of the National Virtual Simulation Experimental Teaching Project in 2018,educational institutions in China have recognized the significance of virtual simulation technology in reforming traditional teaching methods and fostering innovative talent cultivation models.Within the realm of higher education in China,motivating students to sustain their utilization of Virtual Simulation Learning Systems(VSLSs)has become a significant challenge.This article builds upon an assessment of the development status of VSLSs in Chinese higher education and draws upon previous studies to construct a model comprising three dimensions:perceived quality,perceived value,and social influence,with the aim of predicting students’enduring willingness to engage with VSLSs.To achieve this objective,a structural modeling analysis approach is employed to explore the interrelationships among the constructs under investigation,while a survey questionnaire is utilized to collect relevant data.The sample population consists of 274 college students from diverse disciplinary fields in China,including Science,Technology,Engineering,and Mathematics(STEM)and Humanities,Arts,and Social Sciences(HASS).The findings reveal that perceived value significantly influences students’willingness to participate,with perceived benefits exerting a greater impact than perceived costs.Furthermore,the overall quality of the VSLSs,encompassing aspects such as software quality,instructional design quality,and virtual simulation quality,holds substantial influence over students’perceived value.Additionally,societal factors such as course scheduling and recommendations from teachers exhibit a positive impact on students’intention to continue using VSLSs.Building upon these findings,the article presents relevant recommendations aimed at enhancing students’sustained utilization of VSLSs.展开更多
The design and high-throughput screening of materials using machine-learning assisted quantummechanical simulations typically requires the existence of a very large data set,often generated from simulations at a high ...The design and high-throughput screening of materials using machine-learning assisted quantummechanical simulations typically requires the existence of a very large data set,often generated from simulations at a high level of theory or fidelity.Asingle simulation at high fidelity can take on the order of days for a complex molecule.Thus,although machine learning surrogate simulations seem promising at first glance,generation of the training data can defeat the original purpose.For this reason,the use of machine learning to screen or design materials remains elusive for many important applications.In this paper we introduce a new multi-fidelity approach based on a dual graph embedding to extract features that are placed inside a nonlinear multi-step autoregressive model.Experiments on five benchmark problems,with 14 different quantities and 27 different levels of theory,demonstrate the generalizability and high accuracy of the approach.It typically requires a few 10s to a few 1000’s of high-fidelity training points,which is several orders of magnitude lower than direct ML methods,and can be up to two orders of magnitude lower than other multi-fidelity methods.Furthermore,we develop a new benchmark data set for 860 benzoquinone molecules with up to 14 atoms,containing energy,HOMO,LUMO and dipole moment values at four levels of theory,up to coupled cluster with singles and doubles.展开更多
Simulation based structural reliability analysis suffers from a heavy computational burden, as each sample needs to be evaluated on the performance function, where structural analysis is performed. To alleviate the co...Simulation based structural reliability analysis suffers from a heavy computational burden, as each sample needs to be evaluated on the performance function, where structural analysis is performed. To alleviate the computational burden, related research focuses mainly on reduction of samples and application of surrogate model, which substitutes the performance function. However,the reduction of samples is achieved commonly at the expense of loss of robustness, and the construction of surrogate model is computationally expensive. In view of this, this paper presents a robust and efficient method in the same direction. The present method uses radial-based importance sampling (RBIS) to reduce samples without loss of robustness. Importantly, Kriging is fully used to efficiently implement RBIS. It not only serves as a surrogate to classify samples as we all know, but also guides the procedure to determine the optimal radius, with which RBIS would reduce samples to the highest degree. When used as a surrogate, Kriging is established through active learning, where the previously evaluated points to determine the optimal radius are reused. The robustness and efficiency of the present method are validated by five representative examples, where the present method is compared mainly with two fundamental reliability methods based on active learning Kriging.展开更多
文摘In the field of optoelectronics,certain types of data may be difficult to accurately annotate,such as high-resolution optoelectronic imaging or imaging in certain special spectral ranges.Weakly supervised learning can provide a more reliable approach in these situations.Current popular approaches mainly adopt the classification-based class activation maps(CAM)as initial pseudo labels to solve the task.
文摘In this article,reference 43 was incorrectly given as‘Erhard,L.C.,Rohrer,V.,Albe,K.,and Deringer,V.L.,Modelling atomic and nanoscale structure in the silicon-oxygen system through active machine learning,https://arxiv.org/abs/2309.03587(2023)’and should have read‘Erhard,L.C.,Rohrer,J.,Albe,K.et al.Modelling atomic and nanoscale structure in the silicon–oxygen system through active machine learning.Nat Commun 15,1927(2024).https://doi.org/10.1038/s41467-024-45840-9’.
基金the“Doctoral College Advanced Functional Materials-Hierarchical Design of Hybrid Systems DOC 85 doc.funds”funded by the Austrian Science Fund(FWF)and by the Vienna Doctoral School in Physics(VDSP),For Open Access purposes,the author has applied a CC BY public copyright license to any author accepted manuscript version arising from this submission.D.M.and S.P.were supported by the European Union Horizon 2020 research and innovation program under Grant Agreement No.857470the European Regional Development Fund under the program of the Foundation for Polish Science International Research Agenda PLUS,grant No.MAB PLUS/2018/8+2 种基金the initiative of the Ministry of Science and Higher Education’Support for the activities of Centers of Excellence established in Poland under the Horizon 2020 program’under agreement No.MEiN/2023/DIR/3795L.P.and C.F.acknowledge the National Recovery and Resilience Plan(NRRP),Mission 4 Component 2 Investment 1.3-Project NEST(Network 4 Energy Sustainable Transition)of Ministero dell’Universitáe della Ricerca(MUR),funded by the European Union-NextGenerationEUL.L.and C.F.acknowledge the NRRP,CN-HPC grant no.(CUP)J33C22001170001,SPOKE 7,of MUR,funded by the European Union-NextGenerationEU.The computational results were obtained using the Vienna Scientific Cluster(VSC)and the LEONARDO cluster.We acknowledge access to LEONARDO at CINECA,Italy,via an AURELEO(Austrian Users at LEONARDO supercomputer)project.
文摘Understanding and accurately predicting hydrogen diffusion in materials is challenging due to the complex interactions between hydrogen defects and the crystal lattice.These interactions span large length and time scales,making them difficult to address with standard ab-initio techniques.This work addresses this challenge by employing accelerated machine learning(ML)molecular dynamics simulations through active learning.We conduct a comparative study of different ML-based interatomic potential schemes,including VASP,MACE,and CHGNet,utilizing various training strategies such as on-the-fly learning,pre-trained universal models,and fine-tuning.By considering different temperatures and concentration regimes,we obtain hydrogen diffusion coefficients and activation energy values which align remarkably well with experimental results,underlining the efficacy and accuracy of ML-assisted methodologies in the context of diffusive dynamics.Particularly,our procedure significantly reduces the computational effort associated with traditional transition state calculations or ad-hoc designed interatomic potentials.The results highlight the limitations of pre-trained universal solutions for defective materials and how they can be improved by fine-tuning.Specifically,fine-tuning the models on a database produced during on-the-fly training of VASP ML force-field allows the retrieval of DFT-level accuracy at a fraction of the computational cost.
基金the National Social Sciences Found of China’s Major Program“Research on Virtual Reality Media Narrative”(No.21&ZD326).
文摘With the initiation of the National Virtual Simulation Experimental Teaching Project in 2018,educational institutions in China have recognized the significance of virtual simulation technology in reforming traditional teaching methods and fostering innovative talent cultivation models.Within the realm of higher education in China,motivating students to sustain their utilization of Virtual Simulation Learning Systems(VSLSs)has become a significant challenge.This article builds upon an assessment of the development status of VSLSs in Chinese higher education and draws upon previous studies to construct a model comprising three dimensions:perceived quality,perceived value,and social influence,with the aim of predicting students’enduring willingness to engage with VSLSs.To achieve this objective,a structural modeling analysis approach is employed to explore the interrelationships among the constructs under investigation,while a survey questionnaire is utilized to collect relevant data.The sample population consists of 274 college students from diverse disciplinary fields in China,including Science,Technology,Engineering,and Mathematics(STEM)and Humanities,Arts,and Social Sciences(HASS).The findings reveal that perceived value significantly influences students’willingness to participate,with perceived benefits exerting a greater impact than perceived costs.Furthermore,the overall quality of the VSLSs,encompassing aspects such as software quality,instructional design quality,and virtual simulation quality,holds substantial influence over students’perceived value.Additionally,societal factors such as course scheduling and recommendations from teachers exhibit a positive impact on students’intention to continue using VSLSs.Building upon these findings,the article presents relevant recommendations aimed at enhancing students’sustained utilization of VSLSs.
文摘The design and high-throughput screening of materials using machine-learning assisted quantummechanical simulations typically requires the existence of a very large data set,often generated from simulations at a high level of theory or fidelity.Asingle simulation at high fidelity can take on the order of days for a complex molecule.Thus,although machine learning surrogate simulations seem promising at first glance,generation of the training data can defeat the original purpose.For this reason,the use of machine learning to screen or design materials remains elusive for many important applications.In this paper we introduce a new multi-fidelity approach based on a dual graph embedding to extract features that are placed inside a nonlinear multi-step autoregressive model.Experiments on five benchmark problems,with 14 different quantities and 27 different levels of theory,demonstrate the generalizability and high accuracy of the approach.It typically requires a few 10s to a few 1000’s of high-fidelity training points,which is several orders of magnitude lower than direct ML methods,and can be up to two orders of magnitude lower than other multi-fidelity methods.Furthermore,we develop a new benchmark data set for 860 benzoquinone molecules with up to 14 atoms,containing energy,HOMO,LUMO and dipole moment values at four levels of theory,up to coupled cluster with singles and doubles.
基金supported by the National Natural Science Foundation of China (Grant No. 11421091)the Fundamental Research Funds for the Central Universities (Grant No. HIT.MKSTISP.2016 09)
文摘Simulation based structural reliability analysis suffers from a heavy computational burden, as each sample needs to be evaluated on the performance function, where structural analysis is performed. To alleviate the computational burden, related research focuses mainly on reduction of samples and application of surrogate model, which substitutes the performance function. However,the reduction of samples is achieved commonly at the expense of loss of robustness, and the construction of surrogate model is computationally expensive. In view of this, this paper presents a robust and efficient method in the same direction. The present method uses radial-based importance sampling (RBIS) to reduce samples without loss of robustness. Importantly, Kriging is fully used to efficiently implement RBIS. It not only serves as a surrogate to classify samples as we all know, but also guides the procedure to determine the optimal radius, with which RBIS would reduce samples to the highest degree. When used as a surrogate, Kriging is established through active learning, where the previously evaluated points to determine the optimal radius are reused. The robustness and efficiency of the present method are validated by five representative examples, where the present method is compared mainly with two fundamental reliability methods based on active learning Kriging.
